The goal of this application is to identify the key cellular and molecular steps for proper dentin?enamel junction (DEJ) formation. Prior to mineral formation, ameloblasts and odontoblasts are separated by a base- ment membrane. Cell signaling induces the ameloblasts to degrade the basement membrane, move into the rough surface of the mineralizing dentin and secrete enamel matrix proteins that initiate enamel formation. As ameloblasts and odontoblast cell bodies move away from each other when their respective tissues thicken, the odontoblasts maintain functional cell processes within tubules that penetrate the entire width of the dentin. There- fore, the odontoblastic processes are left near the basement membrane. Matrix metalloproteinase-20 (MMP20) is expressed and secreted by both odontoblasts and ameloblasts. MMP20 has preferential (primary) and non- preferential (secondary) enamel matrix cleavage sites and the cleavage site sequential order is essential for proper enamel formation. MMP20 also cleaves cadherin ectodomains, which releases ?-catenin from the inner surface of the cell membrane. Strikingly, MMP20 ablation results in brittle enamel that falls away in sheets from the underlying dentin. Conversely, over-expression of Mmp20 in mice results in significantly softer than normal enamel that contains a cell infiltrate, which is attributed to increased ?-catenin release and signaling. However, the underlying mechanisms for how the DEJ normally develops into such a tough and resilient structure, and how MMP20 regulates this process, remain unknown. This lack of knowledge presents a critical obstacle to our understanding of tooth development. When we establish how MMP20 proteolytic processing supports DEJ for- mation, it will lay the necessary foundation for future steps focused on designing highly effective biomimetic adhesive materials. We will determine the mechanism by which Mmp20 ablation causes disruption of the DEJ (Aim 1). We hypothesize that in normal development, MMP20 contributes to degradation of the basement mem- brane and cleaves enamel matrix proteins, but in the absence of MMP20, this process is compromised so base- ment membrane and matrix proteins remain intact and this may form a barrier that inhibits dentin?enamel ad- herence. We will determine the impact of MMP20 over-expression on the DEJ (Aim 2). We hypothesize that basement membrane hydrolysis and the cleavage order of enamel matrix proteins are required for proper DEJ formation, and that MMP20 over-expression will cause the sites that are normally cleaved in series to be cleaved all at once, resulting in a compromised DEJ, and that too much MMP20 also alters cell signaling. We will deter- mine the extent to which the source of Mmp20 impacts formation of the DEJ (Aim 3). Although both ameloblasts and odontoblasts secrete MMP20, we hypothesize that Mmp20 expression in odontoblasts, which have cell processes near the DEJ, will rescue DEJ formation in KRT14-Cre-Mmp20fl/fl mice. The results are expected to have an important positive impact because they fundamentally advance the field of dental research by defining in molecular terms how enamel and dentin bind together with such strength.